FC proxmox-wazuh-enroll
Install and enroll the Wazuh agent inside Windows VMs hosted on Proxmox VE using QEMU guest agent execution (qm guest exec or Proxmox REST API), with no WinRM, SSH, RDP, GPO, or AD credentials required. Use this skill whenever the user asks to get a Windows Proxmox guest, domain controller, file server, or group of Windows VMs reporting to Wazuh; install, push, roll out, onboard, or enroll the Wazuh agent; fix a just-enrolled Windows Proxmox VM stuck at never_connected; or keep registration secrets off command lines while deploying via guest agent. This skill performs active in-guest deployment and verification against the Wazuh manager. Do not use for read-only Wazuh dashboard, CVE, alert, or coverage checks; Linux, LXC, appliance, or syslog enrollment; generic Proxmox VM creation; generic PowerShell guest exec; or non-Wazuh endpoint agents.
Install and enroll the Wazuh agent inside Windows VMs hosted on Proxmox VE using QEMU guest agent execution (qm guest exec or Proxmox REST API), with no…
As a process C 55/100 · Has gaps — weak spots: result and completion, inputs and preconditions, failures and branches
What is at stake
The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.
The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
How to improve
- Remove the critical guard findings (secrets, dangerous commands, hidden instructions): while they stand the skill is blocked and cannot grade above F.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 3
-
critical Dangerous commands
cmd-encoded-execreferences/qga-exec.md:22Executes a base64/encoded payload (quoted — discussed, not commanded)"\$d=[Text.Encoding]::UT…ing([Convert]::Fr…ing('$b64')); Invoke-Expression \$d" 2>&1quoted
Medium and low: 2
-
medium Exfiltration
net-credential-usereferences/qga-exec.md:30Credential used in a network call (verify the destination is the intended service)PID=$(curl -sk -H "Authorization: PVEAPIToken=$TOKENID=$SECRET" \
-
medium Exfiltration
net-credential-usereferences/qga-exec.md:35Credential used in a network call (verify the destination is the intended service)curl -sk -H "Authorization: PVEAPIToken=$TOKENID=$SECRET" \
Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 55/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 30Running it twice. 3 mutating operations with no state check
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 100Steps. 9 steps
- 100When it triggers. States when to use and when not to
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 690 tokens
- 100Progress reporting. Reports progress
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +3Description length 854: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +4No input/output examples
- +2Single-language instructions
- +4Description says when NOT to use the skill
- +4Structure: 6 headings
- +3Step-by-step instructions: 9 items
- +4Reference files are cited in the instructions (2 of 2)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 86.